Browse Prior Art Database

Sharing/Unsharing Social Media Content Based on Analyzing Sentiment Directed to Specific Users within that Content

IP.com Disclosure Number: IPCOM000241336D
Publication Date: 2015-Apr-17
Document File: 2 page(s) / 110K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is a system to utilize sentiment analysis of social media posts to determine how and whether the information is shared with all the recipients targeted by that sentiment on each recipient’s feed.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 51% of the total text.

Page 01 of 2

Sharing/Unsharing Social Media Content Based on Analyzing Sentiment Directed to Specific Users within that Content

The sharing controls on social media websites are coarse-grained. If a user publishes a post that makes one group happy but is likely to disparage a second group, then the second group should not receive the post. The problem with current methods is that once someone is included in a circle or a tagged in a post, the person automatically receives the post or data on a stream or timeline. A better approach is to not show

unflattering or negative posts, discreetly show the post to an individual for approval, or skip it altogether.

A method is needed to restrict sharing within a circle/network based on the content of the post.

The core idea of this solution is to utilize sentiment analysis of social media posts to determine how and whether the information is shared with all the recipients targeted by that sentiment on each recipient's feed. In some instances, based on preference or sentiment level, the post can be completely excluded. In other instances, the post can be shared as a private mail message. In still other instances, the post can be seen as a flagged entry for investigation depending on the user's preferences associated with the sentiment. Further, depending on the user's current social state, the system can delay certain negative sentiment messages and move those messages to a "holding" area from where the user can review the messages later.

The proposed system uses sentiment analysis of a social media post to determine positive/negative sharing filters. The sentiment analysis considers the sentiment directed toward a person or group of people by weighting the positive/negative sentiment terms near those people. The analysis can determine that a post is largely of positive sentiment while having localized negative sentiment toward one or more people. In this case, the system either flags the post for review or simply does not share it with the person to whom the negative sentiment was directed.

Example post:


"Had a great time at the mall this evening with @User A @User B, and @User C. Can't believe @User D was acting like a jerk! Oh well, lots of fun with my friends!" (Sentiment: bold-positive and underline-negative)

Typical sentiment analysis tools would rate this whole post as positive sentiment or 3/4 positive. However, form User D's perspective, this is a 100% negative sentiment post; therefore, the system does not share the post with User D.

In a preferred embodiment, when a post is made on social media, the system scans the post for positive and negative sentiment. The...